Recombinant adenoviruses have proved to be extremely versatile delivery vehicles for in vivo gene therapy. Their natural tropism to airway epithelial cells and ability to transduce genes into nondividing targets make them particularly attractive for the development of in vivo gene therapies for CF. Preliminary studies performed in several preclinical and clinical models have confirmed the efficiency with which recombinant adenoviruses transfer genes to airway epithelial cells in vivo. In most situations, however, expression is transient and associated with the development of inflammation at the site of transgene expression. We have proposed an hypothesis to explain these important limitations that is based on the expression of viral proteins in cells harboring first generation recombinant adenoviral genomes. This hypothesis predicts that viral proteins are presented to the immune system of the recipient leading to destructive cellular immune responses which destroy the genetically modified cells and repopulate the lung with nontransgene containing cells. Evidence to support this hypothesis has been derived from a series of experiments performed in mice and cotton rats. This component of the SCOR will define the immune responses responsible for target cell destruction and pulmonary inflammation. This hypothesis importantly predicts that recombinant adenoviral technology can be improved if the virus is modified to minimize viral protein expression. The second major component of this project is the development of second generation recombinant adenoviruses in which viral protein expression is diminished or ablated. This will be accomplished through the use of temperature-sensitive mutations and deletion of functions that can be supplied in trans in the context of improved packaging cell lines. Specific modifications of endogenous viral promoters that limit basal expression will also be explored. The development of vectors will be focused as we learn more about the specific components of the adenovirus to which the immune responses are directed. Finally, second generation adenoviruses will be characterized in rodent models including mice and cotton rats to evaluate for diminished inflammation and improved transgene expression.